[visionlist] Call for participants: 2023 Unconstrained Ear Recognition Challenge (@IEEE IJCB 2023)

Vitomir Struc vitomir.struc at fe.uni-lj.si
Mon Apr 17 15:36:13 -04 2023


CALL FOR PARTICIPANTS - UERC 2023

**********************************
The Unconstrained Ear Recognition Challenge 2023 (UERC)

Held in conjunction with IEEE IJCB 2023
https://ijcb2023.ieee-biometrics.org/

Important dates: Registration is open now
UERC 2023 Website: http://awe.fri.uni-lj.si/uerc.html
***********************************


Ear recognition is an active area of research within the biometric 
community. However, the work in this field has long focused on maximizing 
raw recognition performance, while other aspects critical for the 
deployment of biometrics recognition techniques in practice have largely 
been ignored. One such example is demographic bias. Modern ear 
recognition approaches are not only expected to be highly efficient when 
recognizing individuals, but also to be equally fair in their decisions, 
regardless of the demographic characteristics of the subjects, e.g., 
gender or ethnicity. The 2023 Unconstrained Ear Recognition Challenge 
will, therefore, investigate the performance as well as demographic bias 
of existing ear recognition solutions and promote research into 
bias-mitigation mechanisms that have minimal impact on the recognition 
performance.

Understanding demographic bias is important because it can help to 
identify and mitigate inaccuracies and errors in biometric systems, 
prevent the development of discriminatory systems, and even inform the 
public policies and regulations related to biometric systems. Research 
related to demographic bias in ear recognition techniques can help 
promote the development of more accurate, fair, and just ear recognition 
systems that are less likely to produce errors or false positives for 
certain groups of people and protect individual's rights and interests.

To promote research in the bias-aware ear recognition, the Unconstrained 
Ear Recognition Challenge (UERC) 2023 will bring together researchers 
working in the field of ear recognition and benchmark existing and new 
algorithms on a common dataset and under a predefined experimental protocol.


*** How to Participate? ***
To participate in UERC 2023 fill out the following registration form: 
https://docs.google.com/forms/d/e/1FAIpQLSdrXKDcor2MRsvOsdAkjptcCJSZ40NbLHo4Tc5GjP7-pzRGMQ/viewform


*** Execution ***
UERC 2023 will be organized as a two-track competition, where each track 
will be focused on one specific goal. Participants will be free to enter 
only a single track or compete in both. For each track, a dataset, 
evaluation tool written in Python, and baseline models in Python will be 
made available. A detailed description of the two tracks is given below:

Track 1: Fair Ear Recognition. The first UERC 2023 evaluation track will 
collect ear recognition models and score their performance on ear images 
captured in unconstrained environments. Here, the performance indicators 
will include both, a measure of recognition performance as well as an 
estimate of the exhibited demographic bias. Both recognition and bias 
scores will contribute to the overall ranking. Participants will be free 
to develop any type of model to maximize performance, while minimizing 
bias. The final submission for this track will include a working 
solution (source code or compiled binary), which the organizers will run 
to evaluate the performance on a sequestered test data.

Track 2: Bias Mitigation. The second UERC 2023 track will address bias 
mitigation strategies explicitly. Here, a baseline ResNet model (written 
in Python) will be made available to the participants and the goal will 
be to design bias mitigation schemes that reduce the initial bias of the 
models without adversely affecting performance. Such schemes may include 
additional model blocks and network components, normalization layers, 
knowledge infusion mechanisms, score normalization procedures, image 
preprocessing approaches and any other solution capable of reducing bias 
of the predefined base model. Similarly to the first track, participants 
will have to submit a working solution that the organizers will evaluate 
on the sequestered test data.


*** Summary paper and co-authorship ***
The results of UERC 2023 will be published in the IJCB conference paper 
authored jointly by all participants of the challenge.


*** Organizers ***
+ Asst. Prof. Žiga Emeršič, University of Ljubljana, Faculty of Computer 
and Information Science, Slovenia
+ Prof. Hazım Kemal Ekenel, Istanbul Technical University, Department of 
Computer Engineering, Turkey
+ Prof. Guillermo Camara-Chavez, Federal University of Ouro Preto, Brazil
+ Prof. Peter Peer, University of Ljubljana, Faculty of Computer and 
Information Science, Slovenia
+ Prof. Vitomir Štruc, University of Ljubljana, Faculty of Electrical 
Engineering, Slovenia, EU

*** Timeline ***
+ February 14th: Promotion of the competition, website draft, 
registration opens.
+ February 15th: Kick-off of the competition: data, toolkit and 
instructions made available on UERC  website.
+ April 15th: Possible interim ranking.
+ May 1st: Registration closes, end of the competition.
+ May 15th: Summary paper submission.




More information about the visionlist mailing list